Nanotube-based chemical sensors are an emerging class of devices that are able to detect analytes for applications related to human health and security. Current methods to fabricate these devices involve the introduction of carbon nanotubes onto a surface without any methods to facilitate their alignment, and thus networks of randomly oriented nanotubes are formed. Block copolymers, where two different types of polymers are tethered together, are known to assemble into well-ordered morphologies which can be changed by varying the volume fraction of each type of polymer. Thin films of these polymers can be formed on a surface with domains of controllable size. Previously, metals have been deposited on block copolymer thin films to form metal nanowires in an elegant, bottom-up approach. The research outlined in this proposal will explore the alignment of single walled carbon nanotubes (SWCNT) on block copolymer thin films with well-ordered morphologies. The SWCNTs used here will either be unfunctionalized or functionalized on the nanotube's sidewall and end. Devices will be fabricated from the nanotube-substituted polymer thin films and tested for their ability to sense different chemical analytes. Devices that can sense minuscule amounts of gaseous and volatile organic compounds have proven to be important because they can prevent human exposure to hazardous compounds. The research proposed here addresses ways to detect these types of analytes with improved sensitivity derived from the high alignment of the nanotube networks. Similar devices that detect the presence of specific sequences of double stranded DNA (dsDNA) have the potential to be used extensively for disease diagnosis and biowarfare agent detection. While many sensors for the detection of single stranded DNA (ssDNA) have been developed, the direct detection of naturally occurring dsDNA would represent a major improvement in the field. The dsDNA detector developed in this proposal will be evaluated as a tool for the diagnosis of the bacterial infection, Lyme disease, derived from Borrelia burgdorferi bacteria. Popular diagnosis techniques involve identifying the presence of antibodies that the human host raises to B. burgdorferi. The identification of bacterial dsDNA in a human host is a more accurate method to diagnosis Lyme disease, however the current polymerase chain reaction (PCR)-based methods are expensive and require extensive pretreatment. The nanotube-based dsDNA detector proposed here will be able to directly and accurately identify the presence of specific sequences of dsDNA in human blood serum. This could be useful for Lyme disease diagnosis since misdiagnosis can lead to serious long-term side effects and high health care costs. The detector proposed here would represent a powerful tool for the rapid and accurate diagnosis of many diseases caused by bacterial infection.